Latest Posts:

Sorry, no posts matched your criteria.

Follow Us:

Back To Top

JOIN OUR EFFORTS TO

design and scale true in-memory computing

BY LAUNCHING SCIENCE COMPANIES.
Compute architectures have been highly optimised on GPUs and ASICs to accommodate the shift towards specialist processing loads such as neural nets and functions used in financial models. However, the process remains rigid, extremely costly, laden with power-speed tradeoffs and is fixed until the next generation of hardware. This sits awkwardly in a world where new algorithms are discovered at a near monthly cadence and this will only accelerate with time.
Research has touched on escaping the Von Neumman architecture over the years, in-memory compute has moved the memory closer to the processing, non-volatile memory always sits tantalisingly on the horizon and quantum hardware companies promise huge speed ups by recreating gates in the analogue quantum domain.
Until recently it has been infeasible to create architectures that can escape the inherent tradeoffs in CPU, ASIC and FPGA design, both at a materials and software level, whilst maintaining compatibility with existing scalable manufacturing processes.
We have laid the groundwork which suggests that there is potential for speed ups of many orders of magnitude on common computational intensive tasks by progressing all of these elements in parallel. Your mission is to work alongside the existing team to work out how to best align the materials innovation, design, manufacture and software level to bring that technology to life.

 

OUR OFFER
Your role will start as a Founder within DSV under our unique framework for venture creation and transition to becoming part of the founding team of one of the companies which will emerge from it. We are aiming to launch at least three new ventures in this space and you will have a chance to work with us to make that happen (and own an equity stake in each one of them).
There’s up to £500k of investment from DSV alone, and at least £100k in grant funding, plus the further opportunity to carry out 6 figure proof of concept work from our industry partners to be announced soon! In addition to the investment, we also provide a monthly stipend to all our Entrepreneurs in Residence to support their time transitioning into building an early stage company.
No prior idea or IP required.

 

THE IDEAL PROFILE FOR THIS OPPORTUNITY
You bring a mix of technical expertise in a related field (see below) at Master’s level or above and are entrepreneurial at your core, always asking why to get to first principles, highly creative in your solutions and actively enjoy building network, both of peers and more senior, around the problems and opportunities that you go after.

 

Broad experience in semiconductor commercialisation across several parts of the stack:
• Moving from the lab to prototype
• Raising money in semiconductors
• Partnering with fabs
• Market development
Product lead in companies such as AMR, Intel, Broadcom etc.

 

Specific technical edge in at least one of the following:
Circuit design such as:
FPGA / ASIC design and optimisation
Library design of base logic and memory functions
• Memory hierarchy, connectivity architecture and control
• Systems level synthesis and novel methods such as genetic algorithms
• Automated high-level synthesis (HLS) of VHDL / hardware-software co-design and other methods of abstraction
• Power demand modelling and optimisation

 

Memory materials innovation such as:
• Non-volatile memory design broadly
• ReRAM / RRAM / CBRAM, Crossbar, Phase change memory (PCM)
• Oxide performance optimisation around ON/OFF ratio
• Platinum group materials
• Spin Orbit Torque Magnetic (SOT-MRAM)

 

The software level such as:
• Building or managing teams within software that take advantage of hardware acceleration, GPU computing and other parallel computing
• Design or management in low level database-in-memory technology
Compute architectures have been highly optimised on GPUs and ASICs to accommodate the shift towards specialist processing loads such as neural nets and functions used in financial models. However, the process remains rigid, extremely costly, laden with power-speed tradeoffs and is fixed until the next generation of hardware. This sits awkwardly in a world where new algorithms are discovered at a near monthly cadence and this will only accelerate with time.
Research has touched on escaping the Von Neumman architecture over the years, in-memory compute has moved the memory closer to the processing, non-volatile memory always sits tantalisingly on the horizon and quantum hardware companies promise huge speed ups by recreating gates in the analogue quantum domain.
Until recently it has been infeasible to create architectures that can escape the inherent tradeoffs in CPU, ASIC and FPGA design, both at a materials and software level, whilst maintaining compatibility with existing scalable manufacturing processes.
We have laid the groundwork which suggests that there is potential for speed ups of many orders of magnitude on common computational intensive tasks by progressing all of these elements in parallel. Your mission is to work alongside the existing team to work out how to best align the materials innovation, design, manufacture and software level to bring that technology to life.

 

OUR OFFER
Your role will start as a Founder within DSV under our unique framework for venture creation and transition to becoming part of the founding team of one of the companies which will emerge from it. We are aiming to launch at least three new ventures in this space and you will have a chance to work with us to make that happen (and own an equity stake in each one of them).
There’s up to £500k of investment from DSV alone, and at least £100k in grant funding, plus the further opportunity to carry out 6 figure proof of concept work from our industry partners to be announced soon! In addition to the investment, we also provide a monthly stipend to all our Entrepreneurs in Residence to support their time transitioning into building an early stage company.
No prior idea or IP required.

 

THE IDEAL PROFILE FOR THIS OPPORTUNITY
You bring a mix of technical expertise in a related field (see below) at Master’s level or above and are entrepreneurial at your core, always asking why to get to first principles, highly creative in your solutions and actively enjoy building network, both of peers and more senior, around the problems and opportunities that you go after.

 

Broad experience in semiconductor commercialisation across several parts of the stack:
• Moving from the lab to prototype
• Raising money in semiconductors
• Partnering with fabs
• Market development
Product lead in companies such as AMR, Intel, Broadcom etc.

 

Specific technical edge in at least one of the following:
Circuit design such as:
FPGA / ASIC design and optimisation
Library design of base logic and memory functions
• Memory hierarchy, connectivity architecture and control
• Systems level synthesis and novel methods such as genetic algorithms
• Automated high-level synthesis (HLS) of VHDL / hardware-software co-design and other methods of abstraction
• Power demand modelling and optimisation

 

Memory materials innovation such as:
• Non-volatile memory design broadly
• ReRAM / RRAM / CBRAM, Crossbar, Phase change memory (PCM)
• Oxide performance optimisation around ON/OFF ratio
• Platinum group materials
• Spin Orbit Torque Magnetic (SOT-MRAM)

 

The software level such as:
• Building or managing teams within software that take advantage of hardware acceleration, GPU computing and other parallel computing
• Design or management in low level database-in-memory technology

Interviews ongoing

Interviews ongoing